Childs Nerv Syst (2007) 23:321–326 DOI 10.1007/s00381-006-0212-6

ORIGINAL PAPER

Outcome of hemispheric surgeries for refractory epilepsy in pediatric patients Vera Cristina Terra-Bustamante & Luciana Midori Inuzuka & Regina Maria França Fernandes & Sara Escorsi-Rosset & Lauro Wichert-Ana & Veriano Alexandre, Jr. & Marino M. Bianchin & David Araújo & Antônio Carlos Santos & Ricardo Oliveira dos Santos & Helio Rubens Machado & Américo Ceiki Sakamoto Received: 12 May 2006 / Published online: 7 November 2006 # Springer-Verlag 2006

Abstract Background Hemispheric brain lesions are commonly associated with early onset of catastrophic epilepsies and multiple seizure types. Hemispheric surgery is indicated for patients with unilateral intractable epilepsy. Although described more than 50 years ago, several new techniques for hemispherectomy have only recently been proposed aiming to reduce operatory risks and morbidity. Materials and methods We present the clinical characteristics, presurgical workup, and postoperative outcome of a series of pediatric patients who underwent hemispherectomy for medically intractable epileptic seizures. ThirtyV. C. Terra-Bustamante : L. M. Inuzuka : R. M. F. Fernandes : S. Escorsi-Rosset : L. Wichert-Ana : V. Alexandre,:Jr.I M. M. Bianchin : A. C. Sakamoto Department of Neurology, Psychiatry and Psychology, Ribeirão Preto School of Medicine, University of São Paulo, Ribeirão Preto, São Paulo, Brazil R. Oliveira dos Santos : H. R. Machado Department of Neurosurgery, Ribeirão Preto School of Medicine, University of São Paulo, Ribeirão Preto, São Paulo, Brazil

nine patients with medically intractable epilepsy underwent surgery from 1996 to 2005. Results and discussion We analyzed demographic data, interictal and ictal EEG findings, age at surgery, surgical technique and complications, and postsurgical seizure outcome. There were 74.4% males. Tonic and focal motor seizures occurred in 30.8 and 20.5% of the patients. Most frequent etiologies were Rasmussen encephalitis (30.8%) and malformation of cortical development (23.1%). Postsurgical outcomes were Engel classes I and II for 61.5% of the patients. In general, 89.5% of the patients exhibited at least a 90% reduction in seizure frequency. All patients had acute worsening of hemiparesis after surgery. Basically, two surgical techniques have been employed, both with similar results, although a trend has been noted toward one of the procedures which produced consistently complete disconnection. Patients with hemispheric brain lesions usually have abnormal neurological development and intractable epilepsy. When video-EEG monitoring and magnetic resonance imaging show unilateral disease, the patient may evolve with a good surgical outcome. We showed that a marked reduction in seizure frequency may be achieved, with acceptable neurological impairments.

D. Araújo : A. C. Santos Department of Radiology, Ribeirão Preto School of Medicine, University of São Paulo, Ribeirão Preto, São Paulo, Brazil

Keywords Hemispherotomy . Epilepsy surgery in children . Outcome of hemispheric surgery

H. R. Machado (*) Departmento de Neurologia, Centro de Cirurgia de Epilepsia-CIREP, Campus Universitário, Ribeirão Preto-SP, CEP 14048-900, Brazil e-mail: [email protected]

Introduction Children and adolescents usually present with intractable catastrophic epileptic syndromes, characterized by daily seizures or status epilepticus, progressive neurological

322

deficits, and cognitive impairments. Diffuse unilateral lesions are common in this group of patients and the main etiology are malformations of cortical development (MCD), tumors, hypoxic-ischemic lesions, Rasmussen encephalitis, and Sturge–Weber syndrome. These patients may exhibit early seizure onset and multiple seizure types, including infantile spasms, tonic seizures, tonic–clonic seizures, complex partial seizures, and epilepsia partialis continua. Patients with poor seizure control, hemiparesis, and hemianopsia may be good candidates for hemispheric surgeries [17]. Clinical investigation usually shows multifocal or diffuse ictal and interictal EEG findings, involving even the contralateral hemisphere [2]. Magnetic resonance imaging (MRI) helps to determine laterality, extent, and etiology of lesions. Bilateral lesions are present in a large number of patients, especially those with a history of hypoxemia and infectious disease. Functional magnetic resonance imaging, single-photon emission computed tomography, and positron emission tomography may additionally help to determine the functional integrity of the contralateral hemisphere. Hemispheric surgery has been performed in patients with unilateral intractable epilepsies since 1928 [3]. The surgery is typically indicated if epileptic seizures are proven to originate from the abnormal hemisphere, without evidence of contralateral seizure onset [18]. However, some authors have performed hemispheric surgeries even when there are bilateral but asymmetric lesions and seizures originating predominantly on one side [11]. Since the first surgeries, surgical approaches have been progressively improved to avoid large anatomic resections and to minimize postoperative complications and technical modifications are described in the literature according to extension and etiology of the lesion [17, 20]. Anatomic hemispherectomy was the first technique introduced, but the high incidence of late complications lead surgeons to develop new approaches with preservation of ventricular system as in the hemicorticectomies or disconnective surgeries, as the functional hemispherectomy or the hemispherotomies [1, 17, 20]. The main differences between the new techniques are the amount of cerebral tissue resected and the access utilized. At present, disconnective surgeries have been performed with lower morbidity and mortality rate, with a similar seizure outcome [10]. Rasmussen and others at the Montreal Neurological Institute since 1974 initiated the modern era of disconnective surgeries introducing the concept of functional hemispherectomy [19]. After that real innovative technique, coined “hemispherotomy” was introduced by Delalande et al. [4] followed by Villemure and Mascott [19], and many others proposing useful variations in this technique [16]. We aimed to analyze retrospectively in this study the clinical features, the presurgical workup data, the surgical techniques, and the postoperative outcome of a series of

Childs Nerv Syst (2007) 23:321–326

pediatric patients who underwent hemispheric surgeries for medically intractable epileptic seizures in one single center.

Materials and methods We included 39 patients aged up to 18 years, presenting with medically intractable hemispheric epilepsy who underwent surgery from 1996 to 2005. They were assessed at the Ribeirão Preto Epilepsy Surgery Program (CIREP) using previously described presurgical workup, approved by the Ethics Committee of our Institution. The following variables were analyzed: (1) sex, (2) age at epilepsy onset, (3) seizure semiology, (4) seizure frequency, (5) interictal and ictal EEG findings, (6) etiology of epilepsy, (7) age at surgery, (8) surgical technique, (9) surgical complications, (10) duration of follow-up, and (11) postsurgical outcome. Presurgical evaluations included a detailed clinical history and neurological examination, interictal scalp EEG, interictal and ictal video-EEG monitoring (videoEEG), structural and functional imaging, and neuropsychological testing whenever possible. Structural neuroimaging included high resolution 1.5 T Siemens Vision MRI with the following sequences: 3D volume fast-field echo T1weighted, coronal and axial turbo spin echo (TSE) T2weighted, coronal and axial TSE fluid-attenuated inversion recovery, coronal TSE inversion recovery T1-weighted, and sagittal TSE T1-weighted. Patients with suspicion of vascular lesions underwent gadolinium injection. VideoEEG was recorded with a Vangard digital EEG system (Cleveland, OH, USA). EEG samples for interictal spike counts were saved automatically at a rate of 5-min samples per hour. Ictal events were saved by patients, parents, or EEG technicians and were classified according to literature. Interictal EEG was classified in unilateral (90–100% of spikes at one side) or bilateral independent (less than 90% of spikes at one side) or diffuse. Ictal EEG was classified as unilateral, bilateral independent, diffuse, and not available. Surgical technique and outcome Surgery was performed after assigned informed consent by relatives. We have utilized mainly two techniques, although in a few cases other techniques have been employed as anatomic hemispherectomy and hemidecortication (type 3). Although included in our statistics, these cases were not considered for comparison between different techniques. The procedures are described as follows: Type 1: Peri-insular linear corticectomy (preserving the operculum) until the lateral ventricle is opened, from the frontal to the temporal horn, followed

Childs Nerv Syst (2007) 23:321–326

by section of the frontal horizontal fibers, complete callosotomy, section of the fornix, temporal lobectomy (not in all cases), and amygdalohippocampectomy. Type 2: More recently resection of the frontal and parietal operculum, section of the frontal horizontal fibers, complete callosotomy, section of the fornix, amygdalohippocampectomy, and resection of the cortex of the insula. Surgery was considered complete when postopperative MRI showed complete disconnection of all the anatomic structures described in [1] and incomplete when some degree of connection was observed in the frontal region, corpus callosum, insula, and mesial structures of the temporal lobe. After surgery, all patients were evaluated by experienced neurologists at 1, 6, and 12 months, followed additional assessments at 6 months intervals. Patients that were lost to follow-up were contacted by phone to assess surgical outcome in respect to seizure control, neurological, and cognitive development. Postsurgical antiepileptic drug therapy (AED) was maintained unchanged for at least 2 years in patients with monotherapy, and, whenever possible, reduced or converted to monotherapy for patients under AED polytherapy. Seizure outcome was categorized according to Engel’s classification, 1993 [8]. Statistical analysis Fisher’s exact test and the nonparametric Mann–Whitney U test were used for statistical analysis, with the level of significance set at p<0.05.

Results We analyzed 39 patients (28 M, 11 F) with medically intractable epilepsy who underwent epilepsy surgery, with a mean age of 7.5 years (range 2 months to 17 years). The clinical and surgical data are shown in Table 1. All patients had preoperative hemiparesis and 76.5% of the patients had developmental delay. Table 2 summarizes the demographic data. The median age at epilepsy onset was 3.5 years for patients with good outcome (Engel I and II) and 3.3 years for patients with poor outcome (Engel III/IV). There were no statistical differences when comparing patients with good or poor outcome with respect to age of epilepsy, age at surgery, follow-up, duration of epilepsy, and time for seizure recurrence. There was no correlation between sex, history of status epilepticus, side of surgery, surgical technique, and etiologies regarding seizure outcome.

323

In the EEG analysis we observed a good correlation between interictal and ictal abnormalities: 73.7% of the patients with unilateral interictal EEG had also unilateral ictal EEG, and 93.3% of the patients with bilateral interictal EEG had bilateral ictal EEG (p=0.0001). There were no correlations between unilateral interictal or ictal EEG and seizure outcome (p=0.29 and p=0.48, respectively). The most frequent seizure types were tonic seizure (30.8% of the patients) and focal motor seizures (20.5% of the patients), followed by epilepsia partialis continua (17.9% of the patients), and complex partial seizures (10.3% of the patients). Surgical techniques were previously described and only two different techniques were considered in this study for comparison. There were no statistically significant differences considering the outcome for patients submitted to the two main surgical techniques (p=0.14). In 19 patients, postopperative MRI was done (9 patients with surgery type 1 and 10 patients with surgery type 2). A complete disconnection was observed in all patients submitted to type 1 technique and in seven patients with type 2 disconnection. This data was statistically significant (p=0.002). The surgical specimens revealed Rasmussen encephalitis (30.8%), malformation of cortical development (23.1%), porencephalic lesions (20.5%), gliosis (15.4%), Sturge– Weber syndrome (5.1%), tuberous sclerosis (2.6%), and tumors (2.6%), (Table 3). There was no correlation between seizure semiology, etiology of epilepsy, and age of epilepsy onset. Patients with Rasmussen encephalitis had significantly more epilepsia partialis continua (p<0.0001). A good surgical outcome (Engel I and II) was observed in 61.5% of the patients overall. A difference was observed when considering the pathological findings (Table 3), with a better outcome for patients with tumors, tuberous sclerosis complex, porencephalic lesions, and gliosis (70.6% of good outcome). Patients with MCD and Rasmussen encephalitis had a poor outcome with 55.6 and 58.3%, respectively, with good outcome. In 89.5% of the cases, at least 90% reduction of seizure frequency was achieved and 91.7% of the patients when considering Rasmussen encephalitis. Almost all patients had worsening of their hemiparesis and developed hemianopsy after surgery, but all of them were already able to walk unassisted 1 to 6 months after hemispheric resection. One patient developed disabling dysphasia and one patient died for surgical complications 1 week after surgery. All patients had a subjective improvement in quality of life and had their AEDs reduced at follow-up. Two patients died during follow-up, both of pneumonia and sepsis, 3 and 4 years after surgery, respectively. Postoperative hydrocephalus was observed 6 years after surgery in one patient of this series.

Sex

M

F

F

F M M F M M

M

M M M M F

F M M M M

M M M

M M M M F

M M F

Patient number

1

2

3

4 5 6 7 8 9

10

11 12 13 14 15

16 17 18 19 20

21 22 23

24 25 26 27 28

29 30 31

0 0 2

0 2 1 0 1

0 0 6

5 0 0 0 7

3 1 0 0 3

8

0 0 2 9 0 12

3

9

3

Age of epilepsy onset (y)

Tonic Tonic CPS

CPS Tonic Tonic Versive Focal motor

Tonic Tonic Focal motor

Tonic Tonic Infantile spasms Tonic–CPS EPC

Tonic Tonic Infantile spasms CPS Focal motor

Focal motor

Tonic Focal motor EPC CPS Tonic–CPS EPC

EPC

Focal motor

EPC

Seizure type

8 3 15

0 17 7 4 3

7 3 7

8 0 1 4 14

10 4 2 13 12

15

10 1 4 14 2 13

4

10

6

Age of surgery (y)

Table 1 Demographic data of the 34 patients submitted to hemispheric surgeries

Rasmussen encephalitis Rasmussen encephalitis Rasmussen encephalitis MDC SSW MDC MDC SSW Rasmussen encephalitis Rasmussen encephalitis Gliosis Porencephaly Tuberous sclerosis Gliosis Rasmussen encephalitis Porencephaly Gliosis Porencephaly MDC Rasmussen encephalitis Gliosis Porencephaly Rasmussen encephalitis MDC Porencephaly Gliosis Porencephaly Rasmussen encephalitis MDC Tumor MDC

Pathology

Right Right Right

Right Right Right Left Left

Right Right Left

Left Left Left Right Left

Right Right Left Left Left

Left

Left Left Right Right Left Left

Left

Left

Left

Side

I II I

III I I I I

II IV III

I III II III I

III III I I III

III

I I I I III I

III

III

I

Engel

100 97 100

75 100 100 100 100

90 87.5 99.6

100 90 100 75 100

90 99 100 100 99.2

98.3

100 100 100 100 99.8 100

75

90

100

Seizures reduction (%)

1 2 3 3 1 1

2 1 1 2 3

1 2 3 2 3

1 3 1 1 1 Type 1 Type 1 Type 1

Type Type Type Type Type

Type 1 Type 2 Type 2

Type Type Type Type Type

Type Type Type Type Type

Type 3

Type Type Type Type Type Type

Type 2

Type 2

Type 1

Surgical technique

1 6 4 2 2 1 1 1

– Complete –

1 6 6

– Complete – Complete Complete – – Complete

7 1 7 6 1

2 8 4 3 9 Complete – – Incomplete Complete

Complete Incomplete Complete Complete Complete

6

2 6 10 10 2 3

– Incomplete – – Complete Complete Incomplete

3

6

2

Follow-up (y)

Complete

Incomplete

Complete

PO MRI

324 Childs Nerv Syst (2007) 23:321–326

Childs Nerv Syst (2007) 23:321–326

325

Right 14 Tonic–atypical absence 0 F 39

0 0 6 M F M 36 37 38

2 M 35

7 M 34

0 9 M M 32 33

M Male, F female, y years, PO MRI postoperative magnetic resonance image, EPC epilepsia partialis continua, CPS complex partial seizures, MDC malformation of cortical development, SSW Sturge–Weber syndrome, Type 1 and Type 2: see “Surgical technique and outcome” in Materials and methods section for complete description

0 – Type 1

2 5 2 – Incomplete Complete Type 1 Type 2 Type 1 100 100 100 I III I

Gliosis MDC Rasmussen encephalitis MDC 12 2 6

Left Left Left

4 – Type 2 97.3 I 14

Tonic–atypical absence CPS Focal motor Focal motor

Left

6 Incomplete Type 2 100 I 9 EPC

Right

MDC Rasmussen encephalitis Rasmussen encephalitis Porencephaly 1 12 Tonic EPC

Left Right

III I

95 100

Type 1 Type 3

Complete –

2 10

Table 2 Numerical demographic data with respect to seizure outcome Variable

All cases

Engel I/II Median

Engel III/IV Median

p value

1.7 7.0

1.9 8.0

1.1 4.0

0.62 0.06

3.5 14

3.0 14

5.5 19

0.64 0.11

0

7

0

0.98

Median Age at epilepsy onset Age at surgery Epilepsy duration Follow-up Seizure frequency at VEEG Time of seizure recurrence in months

Discussion This study of pediatric epileptic patients with hemispheric brain lesions and hemispheric surgery showed that surgical outcomes were similar among two different surgical techniques but were different concerning etiological profiles. Our series comprised mainly Rasmussen encephalitis (30.8%), MCD (23.1%), and porencephalic cysts (20.5%) patients. All patients had abnormal neurological development, with mild to severe mental retardation associated with hemiparesis and language deficits. Video-EEG showed a great variety of seizure types with a predominance of epilepsia partialis continua, tonic and partial motor seizures as described by other authors in patients with hemispheric lesions [5, 6, 13]. Except for Rasmussen encephalitis, ictal semiology had no specificity for seizure etiology, being the porencephalic lesions those with the greater variety of seizure types. EEG analysis showed a good correlation between unilateral or bilateral abnormalities for interictal and ictal discharges, showing that both interictal and ictal EEG findings have displayed a somewhat neurophysiological coherence. It is interesting to note that patients with bilateral or diffuse EEG findings did not have worse outcomes, suggesting that these EEG profiles do not preclude epilepsy surgery. Table 3 Pathological data and surgical outcome of all patients Pathology

Rasmussen encephalitis MCD Porencephalic cysts Gliosis Sturge–Weber syndrome Tuberous sclerosis Total

Good outcome

Poor outcome

p value

Total (%)

7 5 5 4 1

5 4 2 2 1

0.72 0.70 1.00 1.00 1.00

12 (31.6) 9 (23.7) 7 (18.4) 6 (15.8) 2 (7.0)

1 24

0 14

1.00 –

1 (3.5) 38

MCD Malformations of cortical development

326

Good surgical outcome for hemispheric lesions range from 54 to 69% in the literature [7, 14, 21]. In the literature, several techniques have been used according to the surgeons’ choice and experience, the extension of the lesion, and anatomical findings. In one hand, patients with porencephalic lesions had larger ventricules and less tissue resection is necessary to achieve a complete disconnection. In the other hand, patients with hemimegalencephaly need more tissue resection to permit a complete disconnection. Distortion of the brain may preclude complete disconnection in these cases. Nevertheless, surgical outcome was similar among different hemispherectomy techniques used in this group of patients, even when an incomplete resection was done suggesting that complete resection may not be the only factor associated with good outcome. Otherwise, we observed that surgical outcome was different concerning the etiological factors, with a better outcome for patients with porencephalic lesions, gliosis, tumors, tuberous sclerosis complex, and Sturge–Weber syndrome. A worst outcome was observed in patients with Rasmussen encephalitis and MDC, although in Rasmussen encephalitis group, four patients with Engel III class remained with only nondisabling minor facial jerking seizures. Although hemispheric surgery commonly produces expected postsurgical deficits, patients have reported a subjective increase in quality of life and sometimes taken off their AEDs [12, 15]. Acute and chronic complications related in the literature included hemosiderosis, acute intracranial hypertension, and death [9]. In our series, we had one patient that died in the first week after surgery of brain edema and herniation. Also, both patients with bilateral hemispheric involvement developed postoperative seizures (patient 10 and patient 37 with Engel III classification) and died during late follow-up, suggesting a more severe disease. In conclusion, seizure outcome after hemispheric surgery in patients with extensive hemispheric brain lesions is highly rewarding and can lead to neurological improvement, even when patients persist with frequent but mild seizures. Outcome depends on the etiological profiles, although other factors may be involved and early surgery is advocated for certain etiologies. Acknowledgements The authors thank the epilepsy fellows, neurology residents, and staff of the Epilepsy Outpatient Clinics and VideoEEG Monitoring Unit of the Hospital das Clínicas da Faculdade de Medicina de Ribeirão Preto/USP. This work was supported by FAPESP, CNPq, and FAEPA.

References 1. Carson BS, Javedan SP, Freeman JM, Vining EP, Zuckerberg AL, Lauer JA, Guarnieri M (1996) Hemispherectomy: a hemidecortication approach and review of 52 cases. J Neurosurg 84(6):903–911

Childs Nerv Syst (2007) 23:321–326 2. Chugani HT, Shewmon DA, Peacock WJ, Shields WD, Mazziotta JC, Phelps ME (1988) Surgical treatment of intractable neonatalonset seizures: the role of positron emission tomography. Neurology 38(8):1178–1188 3. Dandy W (1928) Removal of right cerebral hemisphere for certain tumor with hemiplegia. JAMA 90:823–825 4. Delalande O, Pinard J-M, Basdevant C et al (1992) Hemispherotomy: a new procedure for central disconnection. Epilepsia 33(Suppl 3):99–100 5. Devlin AM, Cross JH, Harkness W, Chong WK, Harding B, Vargha-Khadem F, Neville BG (2003) Clinical outcomes of hemispherectomy for epilepsy in childhood and adolescence. Brain 126(Pt 3):556–566 6. Döring S, Cross H, Boyd S, Harkness W, Neville B (1999) The significance of bilateral EEG abnormalities before and after hemispherectomy in children with unilateral major hemisphere lesions. Epilepsy Res 34(1):65–73 7. Duchowny M, Jayakar P, Resnick T, Harvey AS, Alvarez L, Dean P, Gilman J, Yaylali I, Morrison G, Prats A, Altman N, Birchansky S, Bruce J (1998) Epilepsy surgery in the first three years of life. Epilepsia 39(7):737–743 8. Engel J Jr (1993) Protocols for the University of California, Los Angeles. In: Engel J Jr (ed) Surgical treatment of Epilepsies. Raven, New York, pp 743–745 9. Gonzalez-Martinez JA, Gupta A, Kotagal P, Lachhwani D, Wyllie E, Luders HO, Bingaman WE (2005) Hemispherectomy for catastrophic epilepsy in infants. Epilepsia 46(9):1518–1525 10. Jonas R, Nguyen S, Hu B, Asarnow RF, LoPresti C, Curtiss S, de Bode S, Yudovin S, Shields WD, Vinters HV, Mathern GW (2004) Cerebral hemispherectomy: hospital course, seizure, developmental, language, and motor outcomes. Neurology 62 (10):1712–1721 11. Kossoff EH, Vining EP, Pillas DJ, Pyzik PL, Avellino AM, Carson BS, Freeman JM (2003) Hemispherectomy for intractable unihemispheric epilepsy etiology vs outcome. Neurology 61 (7):887–890 12. Maehara T, Shimizu H, Kawai K, Shigetomo R, Tamagawa K, Yamada T, Inoue M (2002) Postoperative development of children after hemispherotomy. Brain Dev 24(3):155–160 13. Ohtsuka Y, Ohno S, Oka E (1999) Electroclinical characteristics of hemimegalencephaly. Pediatr Neurol 20(5):390–393 14. Peacock WJ, Wehby-Grant MC, Shields WD, Shewmon DA, Chugani HT, Sankar R, Vinters HV (1996) Hemispherectomy for intractable seizures in children: a report of 58 cases. Childs Nerv Syst 12(7):376–384 15. Pulsifer MB, Brandt J, Salorio CF, Vining EP, Carson BS, Freeman JM (2004) The cognitive outcome of hemispherectomy in 71 children. Epilepsia 45(3):243–254 16. Shimizu H, Maehara T (2000) Modification of peri-insular hemispherotomy and surgical results. Neurosurgery 47(2):367–373 17. Smith JR, Fountas KN, Lee MR (2005) Hemispherotomy: description of surgical technique. Childs Nerv Syst 21(6):466–472 18. Smith SJ, Andermann F, Villemure JG, Rasmussen TB, Quesney LF (1991) Functional hemispherectomy: EEG findings, spiking from isolated brain postoperatively, and prediction of outcome. Neurology 41(11):1790–1794 19. Villemure JG, Mascott CR (1995) Peri-insular hemispherotomy: surgical principles and anatomy. Neurosurgery 37:975–981 20. Yoshioka H, Hasegawa K, Sakamoto T, Kihara M, Masuda A, Adachi S, Sawada T, Shimizu H (1999) Modified functional hemispherectomy in hemimegalencephaly. Brain Dev 21(2):125– 128 21. Wyllie E, Comair YG, Kotagal P, Bulacio J, Bingaman W, Ruggieri P (1998) Seizure outcome after epilepsy surgery in children and adolescents. Ann Neurol 44(5):740–748